The Matthews correlation coefficient (MCC) should replace the ROC AUC as the standard metric for assessing binary classification.
Davide ChiccoGiuseppe JurmanPublished in: BioData Min. (2023)
Keyphrases
- correlation coefficient
- binary classification
- receiver operating characteristic
- multi class
- support vector
- learning problems
- roc curve
- bipartite ranking
- cost sensitive
- generalization error
- multi class classification
- prediction accuracy
- information gain
- multi label
- ensemble methods
- class distribution
- learning algorithm
- class imbalance
- binary classifiers
- model selection
- error rate
- logistic regression
- principal components
- support vector machine
- classification accuracy
- decision trees
- machine learning
- semi supervised learning
- markov random field
- neural network
- learning tasks
- text categorization